Brouillette 2001.

Methods	Parallel group RCT
Participants	25 children, age 1 to 10 years, with adenotonsillar hyperplasia, signs and symptoms of OSA, and AHI ≥ 1 per hour
Interventions	Fluticasone 50 μg per nostril twice daily for 1 week, followed by 50 μg once daily for 5 weeks; comparison against placebo for 6 weeks
Outcomes	AHI, desaturation index, respiratory arousal index, sleep time with paradoxical movement of chest and abdomen, respiratory rate, heart rate, SpO₂ (mean and minimum), sleep efficiency, adenoidal/nasopharyngeal ratio, minimal airway size, tonsillar size, symptom score; measured after 6 weeks
Notes	Enrolment period: November 1997 to October 1999 Funding sources: The Hospital for Sick Children Foundation and GlaxoWellcome. Declaration of interest: not provided.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization was performed from a table of random numbers."
Allocation concealment (selection bias)	Low risk	"The study drug was dispensed by a pharmacist to an investigator." "Only the study pharmacist was aware of the group assignment."
Blinding (performance bias and detection bias) All outcomes	Low risk	"Fluticasone propionate (Flonase, GlaxoWellcome) and placebo were provided in identical containers by the drug manufacturer. The appearance and smell of the fluticasone and placebo were indistinguishable." "Sleep laboratory personnel, subjects, parents, and physicians were all blinded to group assignment until the entire study was completed and analyzed." "The statistician knew that there were 2 groups but was blinded to which group was placebo and which treatment was used (triple‐blind method)."
Incomplete outcome data (attrition bias) All outcomes	Low risk	All randomised participants finished the study.
Selective reporting (reporting bias)	Low risk	No study protocol or trial registration were available, but outcomes appeared to be complete.
Other bias	Low risk